Introduction
The overrall aim of this project is to try and identify what it means to be “mainstream” both historically, and in the contemporary context. ***
Many artists have been accused and fear being labelled as “mainstream”, but what does that really mean in the context of how streaming websites such as Spotify analyse and characterise music? Through visualisations of a couple of case studies, hopefully spotify’s meainstream sound becomes clearer. ***
Bowie’s Sellout Period visualised
Many fans of David Bowie would say that he sold out(https://time.com/4183623/david-bowie-1980s/) to the masses in the 1980’s, adopting a more radio-friendly sound that would see his hits climbing the charts. 7 of his top 10 chart toppers according to Billboard (https://www.billboard.com/music/david-bowie) were all released between 1983 - 1987. In this study I tried to find a parameter that spotify measures that may reflect the change in his music during this specific period, and see if Bowie’s sellout Period is reflected in the data.
The corpus
The corpus for this study was a playlist curated by Spotify called “This is David Bowie”, which is part of a series of artist focused playlists that spotify creates. The playlsit contains 72 songs that spam his entire career, however there is no information available about how these songs were created, which is a weekness of the corpus.
Findings
Out of all of the parameters that spotify provides, acousticness reflected an interesting result when plotted against the track release dates. As visible in graph 1, there is a clear drop in acousticness around 1980 - 1990. Unfortunately, due to the aesthetic limitations of the graph (and my lack of coding skills) I was unable to determine the exact period in terms of album.track.release_date, however graph 2 visualises the tracks that had a danceability that was less than 0.006. Here, the corpus is reduced to 15 tracks, 7 of which are tracks that were released between 1980 - 1990 which is the highest concentration per decade.
Extra findings
Interestingly, as graph 3 shows, while the songs from the sellout Period are distinguishable by acousticness, when compared with track popularity, they do not stand out of the rest of the corpus. As spotify’s popularity is based only since 2018 (which is when the playlist was created) it seems that modern audiences do not listen to his supposedly more radio-friendly songs any more than the rest of his discography
Graph 3
Coldplay
Can Coldplay’s shift to mainstream sound be seen in Spotify’s data sets
Coldplay is among many bands that many people have grown out of, whether it be because of their gradual shift towards more chart-topping sounds, or simply the social labeling of their music over the years. In this study I tried to use the mean values of various parameters to figure our whether this gradual change can be seen over time.
The Corpus
For this study, I used the mean of all of the parameters offered by spotify for each album that has been released by Coldplay. As I was focusing on albums that were released in a studio setting, I excluded live albums such as “Love in Tokyo” to prevent outliers.
Findings
Unfortunately, none of the parameters apart from those displayed above showed any sign of a trend just based on the album averages, apart from the fact that in general, tempo, valence and energy have been gradual increasing on average over Coldplay’s creative output.